Will Smith
Senior Lecturer in Computer Vision
Department of Computer Science
University of York, UK

Tel: +44 (0)1904 325575
Fax: +44 (0)1904 325599
Mobile: +44 (0)7932 753151
Email: william dot smith at york dot ac dot uk
 
Home Research Publications Teaching Personal


Interests

At the highest level, my interests span computer vision, graphics and machine learning. More specifically, I am interested in the following topics:

These interests ultimately allow me to tackle problems such as:

A wide variety of tools and areas of maths are often useful in my research such as: convex optimisation, nonlinear optimisation, manifold learning, learning/optimisation on manifolds, computational geometry and low level computer vision (e.g. features and correspondence). I am also becoming interested in how higher level image and scene understanding and the context provided by natural language can help solve vision problems.

Current Research Team


Tatsuro Koizumi
MPhil student
Topic: Modelling and learning photometric vision

Chao Zhang
PhD student
Topic: New spaces for shape modelling

Anil Bas
PhD student
Topic: Model-based face analysis in the wild 

Bruce Muller
PhD student
Topic: Geometry meets deep learning for road scene analysis

Sarah Alotaibi
PhD student
Topic: Biophysical face modelling

Dizhong Zhu
PhD student
Topic: Shape from polarisation

Ye yu
PhD student
Topic: Physics-based vision meets deep learning

Former research students and postdocs


Supannee Tanathong
Postdoctoral KTP Associate (November 2014-March 2018)

Current position: R&D, Gaist Solutions Ltd

Hao Guan

PhD Thesis: Local Features, Structure-from-motion and View Synthesis in Spherical Video

After graduating: Computer Vision and Machine Learning Researcher at Shortbite Ltd, UK

Publications from PhD research:
  • Structure-from-motion in Spherical Video using the von Mises-Fisher Distribution. IEEE Transactions on Image Processing (TIP), 2017.
  • BRISKS: Binary Features for Spherical Images on a Geodesic Grid. In Proc. CVPR, 2017.
  • View Stabilisation in Spherical Video. In Proc. OMNIVIS, 2014.
  • Corner detection in spherical images via accelerated segment test on a geodesic grid. In Proc. ISVC, 2011.

Arnaud Dessein
Postdoctoral RA (January 2013-July 2014)

Current position: Postdoc at IMB/LaBRI, Université de Bordeaux, France

Oswald Aldrian

PhD Thesis: Inverse Rendering of Faces with a 3D Morphable Model

After graduating: Postdoc at HFT Stuttgart, Germany

Publications from PhD research:
  • Inverse Rendering of Faces with a 3D Morphable Model. IEEE Trans. Pattern Analysis and Machine Intelligence (PAMI), 2013.
  • Inverse Rendering of Faces on a Cloudy Day. In Proc. ECCV, 2012.
  • Model-based Ambient Occlusion for Inverse Rendering. In Proc. ICIAR, 2012.
  • Inverse Rendering with a Morphable Model: A Multilinear Approach. In Proc. BMVC, 2011.
  • Inverse Rendering in SUV Space with a Linear Texture Model. In Proc. CPCV Workshop at ICCV, 2011.
  • A Linear Approach of 3D Face Shape and Texture Recovery using a 3D Morphable Model. In Proc. BMVC, 2010.
  • Learning the Nature of Generalisation Errors in a 3D Morphable Model. In Proc. ICIP, 2010.

Rui Huang

PhD Thesis: Shape from Shading under Relaxed Assumptions

After graduating: Postdoc at University of Szeged, Hungary

Current position: Assistant Professor at Nanjing University of Posts and Telecommunications

Publications from PhD research:
  • Shape-from-shading under complex natural illumination. In Proc. ICIP, 2011.
  • Photometric Stereo under Low Frequency Environment Illumination. In Proc. ISVC, pages 139-150, 2010.
  • A shape-from-shading framework for satisfying data-closeness and structure-preserving smoothness constraints. In Proc. BMVC, 2009.
  • Structure-preserving Regularisation Constraints for Shape-from-Shading. In Proc. CAIP, 2009.

Ankur Patel
PhD Thesis: 3D Morphable Models: Data Pre-Processing, Statistical Analysis and Fitting

After graduating: Postdoc at Howard Hughes Medical Center (Janelia Farm Research Campus), USA


Current: Senior Data Scientist at ADS Alliance Data Systems Inc., USA

Publications from PhD research:
  • Manifold-based Constraints for Operations in Face Space. Pattern Recognition, 2015.
  • Driving 3D Morphable Models using Shading Cues. Pattern Recognition, 2012.
  • Automated Construction of Low Resolution, Texture-Mapped, Class-Optimal Meshes. IEEE Trans. Visualization and Computer Graphics (TVCG), 2012.
  • Simplification of 3D Morphable Models. In Proc. ICCV, 2011.
  • Exploring the Identity Manifold: Constrained Operations in Face Space. In Proc. ECCV, 2010.
  • 3D Morphable Face Models Revisited. In Proc. CVPR, 2009.
  • Shape-from-shading driven 3D Morphable Models for Illumination Insensitive Face Recognition. In Proc. BMVC, 2009.

 


Abhishek Dutta
MSc by Research Thesis: Face Shape and Reflectance Acquisition using a Multispectral Light Stage

After graduating: PhD student at University of Twente, Netherlands

Current: Senior Software Engineer, TomTom NV, Netherlands

Publications during MSc:
  • Minimal Image Sets for Robust Spherical Gradient Photometric Stereo, Technical Sketch, SIGGRAPH Asia, 2010.

Industry Collaborators

I am very interested in working with forward looking companies who are undertaking exciting research in my areas of interest. I currently collaborate with the following companies:

Academic Collaborators

I have active collaborations with researchers at University of California, San Diego; Aberystwyth University; INRIA Grenoble Rhône-Alpes; Universität Bonn; University of Cambridge; University of Surrey and University of Basel.

Grant Funding

2018-2021 Co-Investigator on Leverhulme Trust project "Archival Polar Photography Unearthing the Forgotten Record of Glacier Change" - Total value: 317k

2014-2017 Academic Supervisor on Innovate UK Knowledge Transfer Partnership with Gaist Solutions Ltd - Total value: 240k

2012-2013 Principal Investigator on DSTL Project "Face Processing in High Volume, Low Quality Data Environments" (DSTLX1000070369) - Total award: 119k

2008-2012 Principal Investigator on EPSRC Project "Combining Model- and Irradiance-based Constraints for Improved Face Shape Recovery and Recognition From Single Images" (EP/F036949/1) - Total award: 127k

An article about this project appeared in The Engineer.

Consultancy

I am available to undertake consultancy work or to provide a professional opinion on research in my areas of interest to the media (an example in New Scientist and Science Magazine).